Columnar thermotropic mesophases formed by dimeric liquid-crystalline ionic liquids exhibiting large mesophase ranges

Article abstract of DOI:10.1039/c1nj20715f

A series of 16 symmetric, dimeric, dicationic liquid-crystalline ionic liquids with mesogenic 3,4,5-tris(alkyloxy)benzyl moieties tethered to two bridged imidazolium cations were designed and synthesised. As a comparison, 12 monocationic imidazolium liqui

Full text of DOI:10.1039/c1nj20715f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 2910–2918
www.rsc.org/njc
PAPER
Columnar thermotropic mesophases formed by dimeric liquid-crystalline
ionic liquids exhibiting large mesophase rangesw
Yanan Gao,* John M. Slattery and Duncan W. Bruce*
Received (in Montpellier, France) 16th August 2011, Accepted 12th September 2011
DOI: 10.1039/c1nj20715f
A series of 16 symmetric, dimeric, dicationic liquid-crystalline ionic liquids with mesogenic
3,4,5-tris(alkyloxy)benzyl moieties tethered to two bridged imidazolium cations were designed
and synthesised. As a comparison, 12 monocationic imidazolium liquid crystals with the
same mesogens were also prepared. The mesomorphic properties of these ionic liquids were
characterised in terms of polarised optical microscopy (POM) and differential scanning
calorimetry (DSC), while thermal stabilities were obtained by thermogravimetric analysis (TGA).
All compounds having chloride (Clꢀ) and tetrafluoroborate (BF4ꢀ) anions exhibit hexagonal
columnar (Colh) liquid crystal mesophases, with the dimeric materials showing mesophases over
an extended temperature range. The effect of the linking chain, alkyl substituents, and anion type
on the thermal properties of these dimers was examined and showed a significant influence. For
ꢀ
example, Colh mesophases were only observed for ionic liquids having Clꢀ and BF4 anions,
whereas systems with the bis(trifluoromethylsulfonyl)imide (Tf2Nꢀ) anion melted directly to
an isotropic liquid. A shorter spacer paired with longer alkoxy chains tends to give rise to
broad-temperature-range Colh ionic liquid crystal phases.
crystallinity at T o100 1C in viologens in 1987,28 hexacatenar
Introduction
pyrilium salts in 1988.29 Furthermore, 1995 also saw the
publication of work describing sub-100 1C mesophases in
N-alkylpyridinium dodecylsulfates30 as well as tetracatenar
complexes of silver(I).31
Materials showing columnar liquid crystal mesophases are of
interest as they offer properties such as improved charge
transportation,1–5 anisotropic ion transportation,6–14 and
photonic15,16 functions. They are also potential components
for organic electronic and optoelectronic devices.17–19 From
this point of view, development of columnar liquid crystals
having a large phase width is of some significance. Typical
discotic liquid-crystalline molecules have a p-electron-rich
aromatic core attached to flexible alkyl chains and columnar
phases are formed via stacking of the disc-like moieties. A
large number of such systems are known and include those
with cores based on triphenylenes, porphyrins, phthalocyanines,
coronenes, and other aromatic molecules.17–23
More recently, organic ionic species such as imidazolium
and ammonium ionic liquids have been used to design func-
tional liquid-crystalline ion conductors.6–14,32,33 Many of these
ionic moieties self-assemble into hexagonal columnar liquid
crystal mesophases.34 One of the goals of molecular electronics
is to develop ion-conductive materials and there can be
advantages when this ion conductivity is anisotropic. Liquid-
crystalline ionic liquids are very promising candidates for the
design of anisotropic, ion-conductive materials because they
have an anisotropic structural organisation and contain ions
as charge carriers. The incorporation of ionic functionality is
also often found to give the liquid crystals a relatively wide
mesophase temperature range.
The first report of liquid-crystalline ionic liquids based on
imidazolium salts is that of Bowlas et al.24 who described
chloride and tetrahalometallate salts of both imidazolium and
pyridinium cations; this work was later followed up by more
comprehensive studies by the Belfast group.25–27 However,
liquid-crystalline ionic liquids of other types pre-date all of this
work, and while it is not intended here to give a comprehensive
overview, attention is drawn (Fig. 1) to reports of liquid
More recently, incorporation of imidazolium cations at the
end of alkylene chains bound to a triphenylene derivative
resulted in stabilisation of a columnar phase and the meso-
phase was maintained over a wide temperature range from 4
to 117 1C.35 A series of fan-shaped molecules, 1-methyl-3-
[3,4,5-tris(alkyloxy)benzyl]imidazolium salts containing anions
such as tetrafluoroborate (BF4ꢀ), hexafluorophosphate (PF6ꢀ),
trifluoromethylsulfonate (CF3SO3ꢀ), and bis(trifluoromethyl-
sulfonyl)imide (Tf2Nꢀ), exhibited hexagonal columnar phases
with a maximum mesophase range of 160 1C.6,34
Department of Chemistry, University of York, Heslington, York,
YO10 5DD, UK. E-mail: yg544@york.ac.uk,
duncan.bruce@york.ac.uk; Fax: (+) 44 1904 322516
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c1nj20715f
c
2910 New J. Chem., 2011, 35, 2910–2918
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 1 Non-imidazolium liquid-crystalline ionic liquids reported in or prior to 1995.28–31
On the other hand, compounds with a symmetric molecular
structure are well known to have a relatively high thermal
stability. The thermal stabilities of the so-called ‘gemini’z
dicationic ionic liquids are found to be greater than those of
most traditional monocationic ionic liquids,36 and as such they
have been used as solvents for high-temperature organic
reactions37 and gas chromatographic stationary phases.38,39
Both symmetric and unsymmetric liquid crystal dimers
composed of two mesogenic groups (identical or different,
respectively) linked via a flexible or rigid spacer have been
reported extensively35,40–48 and their physical properties are
significantly different from those of conventional low-molar-
mass materials.44,49–52 For example, there are very pronounced
odd-even effects that depend upon the parity of the flexible
spacer and, more recently, they have been suggested as materials
with significant flexoelectric response.53
modify dimeric imidazolium ionic liquids to prepare broad-
temperature-range ionic liquid crystal materials that exhibit
fluid ordered states of a columnar nature, maintaining high
ionic conductivities and room-temperature mesophases.
So far, the liquid crystal behaviour of the dimeric imidazolium
dicationic ionic liquids has been scarcely reported. Indeed, there
is rather little literature on flexibly linked, ionic, dimeric liquid
crystals. Initial studies reported examination of the behaviour of
entirely aliphatic diammonium salts55,56 and it was only very
recently that Bara et al.54b took as a template the flexibly linked
bis(imidazolium) salts reported by Anderson et al.36 and simply
(and elegantly) extended the terminal chains to arrive at materials
(Fig. 2) showing smectic A phases, while in addition also using
oligo(ethylene oxides) as spacers, related to studies reported by
Dreja et al.57 Curiously, in some of these reports aspects of the
mesomorphism are interpreted in terms of possible smectic C
phase formation, although it is not at all clear that such a phase
has ever truly been seen in ion-separated ionic liquid crystals.
In ionic systems, the imidazolium cation provides a highly
tunable and modular platform for the design and synthesis of
dimeric cationic liquid crystals which offers excellent structure
control.54 Wishing to introduce ionic moiety into symmetrical
molecular structures, the design strategy employed here is to
z The term dimeric is preferred here both because the term ‘gemini’
properly refers to surfactant systems and also because the dimeric
nature of these materials is much more akin to the flexibly linked
structures found in the study of thermotropic liquid crystals. None-
theless, it is recognised that the early examples of such thermotropic
behaviour originated in derivatives of gemini surfactants.
Fig. 2 General structure of dimeric imidazolium amphiphiles synthe-
sized by Bara et al.54b
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2910–2918 2911

View Article Online
Fig. 3 Structure of triphenylene-imidazole-based ionic dimers reported by Kumar and Gupta.58
A similar idea was then reported by Kumar and Gupta58
General properties of the imidazolium-based ionic liquids
who used the same methylene-bridged bis(imidazolium) core
while attaching triphenylene terminal groups (Fig. 3) to give
materials with Colh phases.
The large number and variety of both symmetric dicationic
dimeric ionic liquids and unsymmetric monocationic ionic
liquids in this study permitted the evaluation of both cationic
structure and anionic effects. Some homologues of 5-n, 6-n and
7-n have been prepared previously (n = 8 and 12)6,34 but new
homologues (n = 10 and 14) were prepared here to provide a
more systematic comparison with the new materials.
In contributing to the study of such dimeric materials, this
work reports the synthesis and thermal properties of a series of
ionic liquid crystal dimers, which have been designed to probe
the relationship between molecular structure and mesophase
behaviour. Structural variables encompassed thꢀe spacer length
(number of repeat units), the anion (Clꢀ, BF4 and Tf2Nꢀ),
and terminal chain length (octyl and dodecyl). For comparison,
12 monocationic imidazolium liquid crystals with the same
mesogenic groups and anions (Clꢀ, BF4ꢀ, and Tf2Nꢀ) were
also prepared.
Polarised optical microscopy was first used to determine the
liquid crystal phase behaviour, but it proved difficult to obtain
good optical textures as most compounds decomposed on or
before clearing. As such, cooling textures were found only for
ꢀ
certain BF4 salts that were found to be thermally stable.
Thus, investigation of dimeric tetrafluoroborates, 3-n,m was
first undertaken and the thermal data are found in Table 1.
All of these salts melted from the crystal phase into a
columnar hexagonal phase, which could be identified by its
characteristic optical texture, an example of which is shown in
Fig. 4. These textures were obtained on slow cooling from the
isotropic liquid in which all homologues appeared stable for
the duration of the experiment and as studied by TGA.
The one exception was 3-12,4, which decomposed/cleared at
255.7 1C and for which no texture could be seen on cooling.
For the chloride salts, textures could not be found on
cooling as all derivatives underwent decomposition and clear-
ing simultaneously; it is not clear which one drives the other.
Therefore the nature of the mesophase that was formed was
determined in two different ways. First, a small portion of the
sample was dissolved in either toluene or dimethyl sulfoxide
(DMSO) and a couple of drops were placed on a microscope
slide on the hot plate, whose temperature was held at 90 1C.
The solvent was allowed to evaporate over a prolonged period
and the texture was observed. An example is given as Fig. 5,
which shows characteristic features of the Colh phase. The
phase identity was further confirmed in several contact experi-
ments, where samples of the BF4 and Cl salts were allowed
to come into contact within their mutual mesophase ranges
but well below the decomposition temperature of the chloride.
Under all conditions, each of these contact preparations
showed the two samples to be continuously miscible, thus
identifying the mesophase as Colh.
Results and discussion
Synthesis of the ionic liquid crystals
The new salts were prepared as follows. Following the procedure
of Kumar and Gupta,58 N-methylimidazole was reacted with the
relevant a,o-dibromoalkylene in the presence of sodium hydride
to give the dimeric bis(imidazole) 1-m; yields ranged from
ca. 60% for 1-4 to Z 95% for 1-6 and 1-8. The dimeric
imidazoles were then further reacted with two molar equiva-
lents of a 3,4,5-trialkoxybenzyl chloride to give the doubly
quaternised dichloride, 2-n,m (m represents the number of
carbon atoms in the bridging methylene chain, while n repre-
sents the number of carbon atoms in the terminal chains).
Metathesis of the anion with AgBF4 or LiTf2N then led to
related salts 3-n,m and 4-n,m.
The 1-(3,4,5-trialkoxybenzyl)-3-methylimidazolium chlorides,
5-n, were prepared by quaternisation of N-methylimidazole while
the corresponding tetrafluoroborates, 6-n, and triflamides, 7-n,
were obtained by metathesis of the chloride using AgBF4 and
LiTf2N, respectively. The compounds were characterised by their
1H and 13C NMR spectra and by elemental analysis. The absence
of halide impurities in the purified BF4 and Tf2N salts was
confirmed by addition of AgNO3 to methanol washes of the
products, which resulted in no observable silver halide precipi-
tates. In total, some twenty-eight compounds were prepared. The
chloride salts were further analysed by Karl-Fischer titration,
which showed a small degree of hydration, consistent with the
results of elemental analysis (Table S1, ESIw).
Examination of the tetrafluoroborate salts of the trialkoxy-
benzylimidazolium cation (6-n) revealed that all homologues
c
2912 New J. Chem., 2011, 35, 2910–2918
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Table 1 Thermal data for symmetric dimeric and unsymmetric ionic
liquids
Compound
Transition
T/1Ca
46.0
241.4
62.3
81.4
236.7
ꢀ20.6
34.4
183.9
49.3
72.3
212.9
35.9
100.5
220.8
36.7
215.8
34.4
237.5
41.4
185.0
98.2
135.3
4.0
255.7
5.9
246.2
36.2
51.0
DH/kJ molꢀ1
2-8,4
Cr–Colh
Colh–I
Cr1–Cr2
Cr2–Colh
Colh–I
Cr1–Cr2
Cr2–Colh
Colh–I
Cr1–Cr2
Cr2–Colh
Colh–I
Cr1–Cr2
Cr2–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr1–Cr2
Cr2–Colh
Colh–I
Cr–I
26.20
1.82
21.93
26.11
1.77
3.36
43.56
1.58
59.18
104.56
2.94
27.22
69.05
2.14
60.99
2.93
12.58
2.43
14.64
2.85
2-8,6
2-8,8
2-12,4
2-12,6
Fig. 4 Optical texture of 3-8,8 at 136.6 1C on cooling.
2-12,8
3-8,4
3-8,6
3-8,8
37.99
2.28
18.89
1.68
7.88
3-12,4
3-12,6
3-12,8
3.90
17.01
12.11
3.88
213.2
42.5
4-8,4
4-8,8
4-12,4
4-12-8
64.52
Cr–I
Cr–I
Cr1–Cr2
Cr2–I
oꢀ40.0
52.8
26.6
116.83
67.94
14.16
38.94
0.95
58.89
1.18
89.43
0.81
40.6
52.2
185.7
64.1
198.7
70.1
187.5
80.6
163.1
ꢀ35.1
132.7
46.9
172.7
61.3
184.8
52.6
58.2
71.6
178.3
9.0
Fig. 5 Optical texture of 2-8,4 obtained after prolonged evaporation
of a DMSO solution at 90 1C.
5-8
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
g–Colh
Colh–I
Cr–Colh
Colh–I
Cr–Colh
Colh–I
Cr1–Cr2
Cr2–Cr3
Cr3–Cr4
Colh–I
Cr–I
5-10
5-12
5-14
6-8b
6-10
6-12
6-14
salts, 5-n, had been reported previously6,34 but there was no
mention of any mesomorphic properties. In common with
chlorides 2-n,m, these materials were also unstable thermally
and so no textures could be obtained on cooling from the
isotropic state. Thus, the mesophase was once more identified
in contact preparations with known materials described above
and a Colh phase was found in each case.
121.96
0.98
0.96
40.39
0.75
55.61
0.81
Finally here, it is noted that none of the triflamide salts
(4-n,m nor 7-n) was mesomorphic, attributed to the destabilis-
ing effect of the large anion (large excluded volume), which is
also evident in the rather low melting points, all of which are
below 60 1C. Dependence of liquid crystal properties on anion
size for ionic liquids is a well-described phenomenon.34,59,60
In considering the thermal stability of these materials, the
chlorides and tetrafluoroborates follow the general trends
recorded elsewhere both in simple 1-butyl-3-methylimid-
azolium (bmim) ionic liquids and in non-mesomorphic,36,61
dimeric materials akin to those reported here. In all of the
other studies, triflimides are found to be even more stable, but
given the low melting points of the materials described here, no
comparisons are possible.
67.44
0.96
7-8
7-10
37.90
19.74
21.74
72.25
80.79
Cr1–Cr2
Cr2–I
Cr–I
ꢀ13.4
19.3
39.7
7-12
7-14
Cr–I
51.0
a
b
Temperatures are DSC onset values. g = glass transition.
showed a Colh phase as determined unequivocally by their
optical textures, which could be obtained on cooling from the
isotropic liquid owing to their thermal stability. These salts
mostly melted around 50–60 1C with the exception of 6-8
where it was not possible to observe a melting point as the salt
was already liquid crystalline at room temperature, although
low-temperature DSC measurements did show what appeared
to be a glass transition at ꢀ35 1C. The corresponding chloride
Given the thermal instability of these salts, only limited
structure/property relationships can be extracted. Fig. 6 shows
the clearing points of the tetrafluoroborates 3-n,m as a func-
tion of both spacer and terminal chain length, and these are
plotted against the limits of stability of the corresponding
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2910–2918 2913

View Article Online
It is here that comparison with the monosubstituted imid-
azolium salts becomes of significance, for compounds 6-n can
be regarded as approximating to the terminal ‘half’ of flexible
dimers 3-n,m. Thus, in the ‘monomers’, a combination of the
new data recorded here and those reported previously shows
that as n increases as 8, 10, 12 and 14, the clearing point
increases as 132.7, 172.7, 184.8 and 178.3 1C, respectively.
Thus, between n = 8 and 12 the clearing point increases by
52.1 1C, which is of a similar magnitude to the values found
in 3-n,m, validating the comparison. Thus, the effect of the
terminal chains is not dampened by the flexible spacer.
Other features of the thermal behaviour of these materials
are worthy of note. Thus, they show extraordinarily large
mesomorphic ranges for materials of this type. The largest
absolute range is found for 3-12,4 at 255.7 1C, but this
compound decomposes on clearing and so the largest useable
range is for 3-12,6, one of 246.2 1C. Kinetically, the range of
these materials is further extended as they show a significant
propensity to supercool well below their melting points and
this is something that could readily be exploited more system-
atically in mixtures where true eutectics could be formed to
give even wider thermodynamic ranges.
Fig. 6 Plot of clearing point vs. compound—BF4 salts (3-n,m) in
lighter shading and Cl (2-n,m) salts in darker shading. ‘Clearing points’
for the Cl salts are in fact the highest temperature at which a phase is
seen and it is not possible reliably to know if clearing drives decom-
position or vice versa. Note that the ordinate starts at 120 1C, which is
above the melting point of all the samples in the plot.
Prior to this report, the largest phase range for columnar
ionic liquid crystals was observed in a tripodal 4-(trialkoxy-
phenyl)pyridinium salt with PF6ꢀ as the anion, which showed
a rectangular columnar phase from 24 to 245 1C.14 Other wide
ranges were found in triphenylene derivatives bearing tetra-
decyl chains with imidazolium termini, which formed wide-
range cubic mesophases60 and the pyrylium salts mentioned
earlier where the mesophase extended from ambient tempera-
ture to over 200 1C.29
As for the dicationic imidazolium materials reported here,
these large-temperature-range ionic liquid crystals undoubtedly
represent a promising new family of media with many exciting
potential applications, in which a large liquid crystal range and
high thermal stability can be exploited.
chlorides, 2-n,m. With both an octyloxy and a dodecyloxy
terminal chain, compounds 3-n,m show a monotonic decrease
in mesophase stability as the length of the spacer chain
increases. This is most pronounced for 3-8,m where the clear-
ing point drops from ca 237.5 1C for m = 4 to 135.3 1C for
m = 8—about 50 1C for each pair of methylene groups. For
the series 3-12,m, the effect is less dramatic. The linear
correlation across all three homologues is not evident owing
to the decomposition of 3-12,4 at 255.7 1C, but extrapolating
a line from 3-12,6 and 3-12,8 would imply a clearing point of
282 1C to give a gradient of 34 1C per two methylene groups.
That the clearing point decreases with an increase in the length
of the methylene spacer is entirely consistent with the behaviour
in more conventional flexibly linked dimeric liquid crystals and
is understood in relation to the gradual decoupling of the two
ends as the chain length increases.
Moreover, the dimeric dicationic ionic liquids were found to
exhibit a slightly higher thermal stability than unsymmetric
monocation analogues, and compounds 2-12,4, 3-12,4, 4-12,4,
and unsymmetric 5-12, 6-12, 7-12 are used as an example as
they have the same terminal chains. The thermal decomposi-
tion temperatures determined by TGA for the two series of
compounds are collected in Table 2. Symmetric dicationic
2-12,4 showed a decomposition temperature of 244 1C, while
the corresponding unsymmetric 5-12 decomposed at 200 1C;
The decomposition temperature is 280 1C for 3-12,4 but
decreased to 260 1C for 6-12. A similar trend was also found
Given the lower thermal stability of the chlorides 2-n,m, a
broad and meaningful comparison is not possible, particularly
for 2-12,m which have all decomposed by 220 1C. However,
some comparison of 2-8,m and 3-8,m is possible as the
chlorides decompose above the clearing points of the tetra-
fluoroborates. Thus, the columnar phase of the chlorides is in
all cases more stable and for m = 6 and 8, the extent is at least
50 1C. This is consistent with the lower steric demands of
the chloride anion and is accentuated as the linking chain
increases in length, but where this chain is short (m = 4) then
the effect is diminished as clearing points are also elevated
significantly owing to the small value of m.
Table 2 Thermal decomposition temperatures for symmetric dimeric
and unsymmetric ionic liquids
Compound
T/1Ca
2-12,4
3-12,4
4-12,4
5-12
6-12
7-12
244
280
307
200
260
273
The other factor of particular note is the effect of the terminal
alkoxy chain length. Thus, if the extrapolated clearing point
of 3-12,4 is accepted to be 282 1C, then for m = 4, 6 and 8, the
clearing point increases by 45, 61 and 78 1C, respectively as n
increases from 8 to 12—very significant enhancements.
a
Temperatures are TGA onset values.
c
2914 New J. Chem., 2011, 35, 2910–2918
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
by the comparison between 4-12,4 and 7-12, the decomposi-
tion of the former took place at 307 1C, while 273 1C for the
latter. It can be further seen from Table 2 that the triflamide
salts are more stable than BF4 and further Cl salts.
Thermal properties of liquid crystal transitions were measured
using a Mettler Toledo DSC822e differential scanning calori-
meter, equipped with Mettler Toledo TS0801RO sample robot
and calibrated against pure indium metal. Heating and cooling
rates were 10 1C minꢀ1. Optical textures were recorded using
an Olympus BX50 polarising microscope equipped with a
Linkam scientific LTS350 heating stage, Linkam LNP2 cooling
pump, and Linkam TMS92 controller. Thermogravimetric
analysis was performed on a Stanton-Redcroft STA 625 instru-
ment. The heating rate employed was 10 1C minꢀ1 from room
temperature (approximately 25 1C).
In addition, the dimeric dicationic ionic liquids have gen-
erally exhibited wider mesophase ranges than the unsymmetric
compounds for Cl salts or BF4 salts. Unsymmetric 5-8 showed
a Colh mesophase of range 133.5 1C, whereas the mesophase
range found for dimeric 2-8,4, 2-8,6 and 2-8,8 are 197.4, 155.7
and 149.5 1C, respectively. Compound 5-12 has a mesophase
range of 117.4 1C, which is lower than the corresponding
symmetric materials 2-12,4 (140.6 1C), 2-12,6 (120.3 1C) and
2-12,8 (178.9 1C). A similar trend can also be found for BF4
salts. Here, the mesophase range of symmetric 3-12,4, 3-12,6
and 3-12,8 are 251.7, 240.3 and 162.2 1C, respectively, while
a range of 123.5 1C was observed for unsymmetric 6-12. The
one exception was 6-8, for which a melting point has not
been recorded (vide supra). However, the inability to cause the
material to crystallise and the observation of a glass transition
at ꢀ35.1 1C suggests a useful range of up to 167.8 1C, although
it is recognised that this is not a true thermodynamic range.
Synthesis
The synthetic pathways used to obtain symmetric 2-n,m;
3-n,m; 4-n,m and unsymmetric 5-n; 6-n; and 7-n are shown in
Scheme 1. Diimidazole compounds (1-m) were first synthesised
from imidazole and the corresponding a,o-dibromoalkanes
according to procedures outlined in the literature,58,62 but with
minor modifications to the purification protocol. The meso-
genic 3,4,5-tris(alkyloxy)benzyl chlorides (n = 8, 10, 12, and
14) were then synthesised by a literature procedure.6 Subse-
quent reaction with 2 equivalents of a 3,4,5-trialkoxybenzyl
chloride produced the symmetric dicationic ionic liquids with
chloride counterions (2-n,m). Ion-exchange with AgBF4 or
LiTf2N afforded the related salts 3-n,m and 4-n,m. In a similar
way, unsymmetric monocationic 5-n compounds were pre-
pared by quaternisation of N-methylimidazole with mesogenic
3,4,5-tris(alkyloxy)benzyl chlorides. 6-n and 7-n are obtained
by metathesis of the chlorides using AgBF4 and LiTf2N,
respectively.
Conclusion
We have demonstrated that the symmetric dicationic imida-
zolium ionic liquids (2-n,m and 3-n,m, n = 8, 12; m = 4, 6, 8)
and unsymmetric monocationic imidazolium ionic liquids (5-n
and 6-n, n = 8, 10, 12, 14) are able to form Colh mesophases.
Their symmetric structure and the presence of the ion moiety
leads to a relatively high thermal stability and wide mesophase
range. Indeed, an extremely large columnar phase width of
more than 200 1C and spanning room temperature was obser-
ved for the first time. Spacer length, alkyl length and the nature
of ions were all found to influence the mesomorphism. Thus,
systems with longer alkyl chain length and short spacer length
were found to have large mesophase ranges. In considering
Synthesis of diimidazole compounds
In a typical procedure, a solution of imidazole (5.0 g, 72.4 mmol)
in DMF (25 cm3) in a round-bottomed flask equipped with
a stirring bar was deaerated under reduced pressure, and the
flask was filled with nitrogen. The deaeration was repeated three
times to remove oxygen from the flask thoroughly. The flask
was cooled down to about 0 1C with an ice bath. Sodium
hydride (1.92 g, 80.0 mmol) was then added slowly into the
reaction solution. After that, dibromoalkane (33.35 mmol) was
added and the resulting mixture was heated at 70 1C for 4 h
with vigorous stirring. When the reaction was over, the crude
products were purified either by crystallisation or column
chromatography, which depends on the properties of the pro-
ducts. The purification processes are as follows: For 1,10-(butane-
1,4-diyl)diimidazole (1-4), after the reaction was finished, the
mixture was cooled down to room temperature and then poured
into water (15 cm3). A white, crystalline solid precipitated from
aqueous solution after water was evaporated in air for at least
24 h. The same crystallisation was performed twice and 3.68 g
(58%) of pure solid product was obtained after dying under
decreased pressure at 60 1C for 24 h. 1H NMR (400 MHz,
CDCl3): d = 7.37 (s, 2H), 7.00 (s, 2H), 6.80 (s, 2H), 3.87 (m, 4H),
1.69 (m, 4H). 13C NMR (100 MHz, CDCl3): d = 136.9, 129.8,
118.5, 46.3, 28.1.
the counter anions, Clꢀ and BF4 were found to stabilise the
ꢀ
formation of mesophases, while the bulky anion, Tf2Nꢀ,
suppressed phase formation. Symmetric, dimeric, dicationic
imidazolium ionic liquid crystals therefore represent a versatile
platform for molecular design. For example, the coupling of
the remarkable mesophase range and the incorporation of ions
into the centre of Colh phase demonstrate the potential for
anisotropic ionic conductivity. Exploring their fundamental
structure-property relationships also give a guide for the design
of high-temperature-range ionic liquid crystals.
Experimental
General procedures and materials
Reagents and solvents were obtained from commercial suppliers
and were used without further purification. The compounds were
1
characterised by H and 13C NMR spectroscopy recorded on
JEOL ECX400, or Bruker AV500 spectrometers. Elemental
analysis was carried out on an Exeter Analytical Inc. CE-400
Elemental Analyser. Mass spectra were obtained on a Bruker
microTOF instrument. The water content in samples was
measured by Karl Fischer (Mettler Toledo DL32 coulometer).
For 1,10-(hexane-1,6-diyl)diimidazole (1-6), the solvent was
removed at 80 1C under decreased pressure after the reaction
was complete. Chloroform (100 cm3) was then added to the
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2910–2918 2915

View Article Online
Scheme 1 Synthesis of dimeric dicationic ionic liquids and unsymmetric monocationic ionic liquids: (i) Br(CH2)mBr, DMF, NaH (ii) toluene
(iii) AgBF4, MeOH or LiTf2N, MeOH.
residual and the resulting mixture was dried over anhydrous
MgSO4, filtered through a pad of celite, and concentrated
under reduced pressure. The crude solid was purified by silica
gel flash-column chromatography eluting with 10% methanol
in ethyl acetate to give 6.90 g (95%) of the pure product as a
crystalline solid. 1H NMR (400 MHz, CDCl3): d = 7.38 (s, 2H),
6.99 (s, 2H), 6.82 (s, 2H), 3.83 (m, 4H), 1.69 (m, 4H), 1.23 (m, 4H).
13C NMR (100 MHz, CDCl3):d = 137.0, 129.4, 118.6, 46.7,
30.8, 26.0.
For 1,10-(octane-1,8-diyl)diimidazole (1-8), after the reac-
tion was over, the mixture was poured into a mixture of ethyl
acetate and water (10 : 1). The organic phase was separated;
the aqueous phase was extracted with ethyl acetate three times.
The combined organic extracts were washed with water and
saturated NaCl solution, respectively. The resulting organic
phase was dried over anhydrous MgSO4, filtered through a
pad of celite, and concentrated under reduced pressure to give
7.96 g (97%) of the pure product as a viscous liquid.y
1H NMR (400 MHz, CDCl3): d = d 7.38 (s, 2H), 6.98
(s, 2H), 6.83 (s, 2H), 3.84 (m, 4H), 1.71 (m, 4H), 1.21 (m, 4H).
13C NMR (100 MHz, CDCl3): d = 136.9, 129.2, 118.7, 46.8,
30.8, 28.7, 26.3.
Synthesis of 1,10-(1,4-butanediyl)bis{3-[3,4,5-tris(octyloxy)benzyl]-
imidazolium} chloride (2-8,4). To a solution of 3,4,5-tris-
(octyloxy)benzyl chloride (10.21 g, 20.0 mmol) in toluene
y One referee pointed out that this compound has previously been
obtained as a solid, but using SciFinder we were unable to find
corroborative evidence. Indeed, leaving the sample in a freezer at
ꢀ20 1C overnight did not cause solidification.
c
2916 New J. Chem., 2011, 35, 2910–2918
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
(10 cm3) under nitrogen was added a solution of 1,10-(butane-
1,4-diyl)diimidazole (1-4) (0.95 g, 5.0 mmol) in MeCN (10 cm3)
with stirring. After the reaction solution was heated at 80 1C for
48 h, the solvents were removed by a rotary evaporator under
reduced pressure. The product was separated and purified by
silica gel flash-column chromatography eluting with CH2Cl2/
MeOH (10 : 1) and dried under reduced pressure at 60 1C for
24 h to give the pure product as a white solid (5.30 g, 87.5%).
1H NMR (400 MHz): d = 10.61 (s, 2H), 7.89 (t, 2H), 7.06 (t, 2H),
6.58 (s, 4H), 5.31 (s, 4H), 4.55 (t, 4H), 3.93–3.90 (m, 12H), 2.18
(t, 4H), 1.79–1.67 (m, 12H), 1.48–1.26 (m, 64H), 0.87–0.84
(m, 18H). 13C NMR (100 MHz): d = 153.9, 138.9, 136.9, 127.8,
123.2, 121.3, 107.6, 73.5, 69.5, 53.9, 49.1, 32.0, 31.9, 30.4, 29.6,
29.5, 29.4, 26.6, 26.2, 26.2, 22.8, 14.2.
139.3, 135.1, 126.7, 123.0, 122.0, 121.4, 118.2, 107.5, 73.6, 69.4,
54.3, 49.4, 32.0, 30.4, 29.6, 29.5, 29.4, 26.7, 26.2, 26.1, 22.8, 14.2.
The synthetic procedures for 4-8,8, 4-12,4 and 4-12,8 and
all unsymmetric monocationic ionic liquids are given in the
Supporting Information, along with data for elemental analysis
for all compounds.
Acknowledgements
Financial support for this research from Royal Society BP
Amoco Fellowship (NF080935) is acknowledged gratefully.
References
1 D. Adam, P. Schuhmacher, J. Simmerer, L. Haussling,
K. Siemensmeyer, K. H. Etzbach, H. Ringsdorf and D. Haarer,
Nature, 1994, 371, 141.
2 T. Kato, N. Mizoshita and K. Kishimoto, Angew. Chem., Int. Ed.,
2006, 45, 38.
A similar synthesis procedure was used to produce 2-8,6,
2-8,8, 2-12,4, 2-12,6 and 2-12,8 and the relevant data are listed
in the Supporting Information.
3 M. Lehmann, G. Kestemont, R. Gomez Aspe, C. Buess-Herman,
M. H. J. Koch, M. G. Debije, J. Piris, Matthijs, P. de Haas,
J. M. Warman, M. D. Watson, V. Lemaur, J. Cornil, Y. H. Geerts,
R. Gearba and D. A. Ivanov, Chem.–Eur. J., 2005, 11, 3349.
Synthesis of 1,10-(1,4-butanediyl)bis{3-[3,4,5-tris(octyloxy)benzyl]-
imidazolium} tetrafluoroborate (3-8,4). To a solution of 2-8,4
(2.42 g, 2.0 mmol) in CH2Cl2 (15 cm3) was added a solution of
silver tetrafluoroborate (AgBF4) (0.86 g, 4.4 mmol) in MeOH
(15 cm3) with stirring at room temperature. The mixture was
stirred at room temperature for 2 h. The insoluble AgCl was
filtered off through a pad of celite by using a suction funnel.
The filtrate was concentrated by using a rotary evaporator.
The crude product was purified by flash-column chromato-
graphy on silica gel (eluent: CH2Cl2/MeOH = 10 : 1) and
dried under reduced pressure at 60 1C for 24 h to give the pure
product as a white solid (2.22 g, 84.4%). 1H NMR (500 MHz):
d = 8.81 (s, 2H), 7.40 (t, 2H), 7.11 (t, 2H), 6.58 (s, 4H), 5.17 (s, 4H),
4.22 (t, 4H), 3.96–3.90 (m, 12H), 1.97 (t, 4H), 1.80–1.68 (m, 12H),
1.48–1.26 (m, 64H), 0.88–0.84 (m, 18H). 13C NMR (100 MHz):
d = 154.0, 139.0, 135.6, 128.9, 122.8, 121.7, 107.5, 73.5, 69.4,
54.2, 49.2, 32.0, 31.9, 30.4, 29.6, 29.5, 29.4, 26.2, 22.8, 14.2.
ESI-MS (methanol, m/z): 1337.9449 (1337.9493 calculated
[M]+ C72H124B2F8N4NaO6).
4 C. D. Simpson, J. Wu, M. D. Watson and K. Mullen, J. Mater.
Chem., 2004, 14, 494.
¨
5 S. Sergeyev, W. Pisula and Y. H. Geerts, Chem. Soc. Rev., 2007,
36, 1902.
6 M. Yoshio, T. Mukai, H. Ohno and T. Kato, J. Am. Chem. Soc.,
2004, 126, 994.
7 M. Yoshio, T. Kagata, K. Hoshino, T. Mukai, H. Ohno and
T. Kato, J. Am. Chem. Soc., 2006, 128, 5570.
8 H. Shimura, M. Yoshio, K. Hoshino, T. Mukai, H. Ohno and T. Kato,
J. Am. Chem. Soc., 2008, 130, 1759.
9 M. Yoshio, T. Mukai, K. Kanie, M. Yoshizawa, H. Ohno and
T. Kato, Adv. Mater., 2002, 14, 351.
10 T. Ichikawa, M. Yoshio, A. Hamasaki, T. Mukai, H. Ohno and
T. Kato, J. Am. Chem. Soc., 2007, 129, 10662.
11 T. Kato, Angew. Chem., Int. Ed., 2010, 49, 7847.
12 T. Kato, Science, 2002, 295, 2414.
13 A. E. Frise, T. Ichikawa, M. Yoshio, H. Ohno, S. V. Dvinskikh,
T. Kato and I. Furo, Chem. Commun., 2010, 46, 728.
14 K. Tanabe, T. Yasuda and T. Kato, Chem. Lett., 2008, 37, 1208.
15 F. Wurthner, C. Thalacker, S. Diele and C. Tschierske, Chem.–Eur. J.,
¨
2001, 7, 2245.
16 M. O’Neill and S. M. Kelly, Adv. Mater., 2003, 15, 1135.
17 S. Laschat, A. Baro, N. Steinke, F. Giesselmann, C. Hgele,
G. Scalia, R. Judele, E. Kapatsina, S. Sauer, A. Schreivogel and
M. Tosoni, Angew. Chem., Int. Ed., 2007, 46, 4832.
18 K. Ichimura, S. Furumi, S. Morino, M. Kidowaki, M. Nakagawa,
M. Ogawa and Y. Nishiura, Adv. Mater., 2000, 12, 950.
19 M. Lehmann, I. Fischbach, H. W. Spiess and H. Meier, J. Am.
Chem. Soc., 2004, 126, 772.
20 Introduction to Liquid Crystals, ed. P. J Collings and M. Hird,
Taylor and Francis, London, 1997.
21 S. Kumar, Liq. Cryst., 2009, 36, 607.
22 H. K. Bisoyi and S. Kumar, Chem. Soc. Rev., 2010, 39, 264.
23 S. Kumar, Chem. Soc. Rev., 2006, 35, 83.
24 C. J. Bowlas, D. W. Bruce and K. R Seddon, Chem. Commun.,
1996, 1625.
25 C. Hardacre, J. D. Holbrey, P. B. McCormac, S. E. J. McMath,
M. Nieuwenhuyzen and K. R. Seddon, J. Mater. Chem., 2001,
11, 346.
26 J. D. Holbrey and K. R. Seddon, J. Chem. Soc., Dalton Trans.,
1999, 2133.
27 C. M. Gordon, J. D. Holbrey, A. R. Kennedy and K. R. Seddon,
J. Mater. Chem., 1998, 8, 2627.
28 K. Yamamura, Y. Okada, S. Ono, K. Kominami and I. Tabushi,
Tetrahedron Lett., 1987, 28, 6475.
Similarly, the synthesis procedure of 3-8,6, 3-8,8, 3-12,4,
3-12,6 and 3-12,8 is shown in Supporting Information.
Synthesis of 1,10-(1,4-butanediyl)bis{3-[3,4,5-tris(octyloxy)benzyl]-
imidazolium} bis(trifluoro-methylsulfonyl)imide (4-8,4). To a
solution of 2-8,4 (0.61 g, 0.5 mmol) in MeOH (50 cm3) was
added a solution of LiTf2N (0.32 g, 1.1 mmol) in MeOH
(15 cm3) with stirring at room temperature. The mixture was
stirred at room temperature for 10 min. After removal of the
solvents at 60 1C under decreased pressure, the residual was
dissolved in CH2Cl2 (10 cm3) and poured into water (5 cm3).
The organic phase was separated, washed with water for 2
times and dried over anhydrous MgSO4, filtered through a
short pad of celite, concentrated by a rotary evaporator. The
crude product was purified by flash-column chromatography
on silica gel (eluent: CH3Cl3/MeOH = 10 : 1) and dried under
decreased pressure at 60 1C for 24 h to give the pure product
as a very viscous liquid (0.61 g, 74.4%). AgBF4 test shows no
1
Clꢀ remaining. H NMR (500 MHz): d = 8.77 (s, 2H), 7.41
29 H. Strzelecka, C. Jallabert and M. Veber, Mol. Cryst., Liq. Cryst.
Inc. Nonlin. Opt., 1988, 156, 355.
30 D. W. Bruce, S. Estdale, D. Guillon and B. Heinrich, Liq. Cryst.,
1995, 19, 301.
(t, 2H), 7.12 (t, 2H), 6.52 (s, 4H), 5.14 (s, 4H), 4.25 (t, 4H),
3.94–3.90 (m, 12H), 2.01 (t, 4H), 1.75–1.68 (m, 12H), 1.46–1.26
(m, 120H), 0.88–0.84 (m, 18H). 13C NMR (100 MHz): d = 154.1,
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2910–2918 2917

View Article Online
31 D. W. Bruce, B. Donnio, D. Guillon, B. Heinrich and M. Ibn-Elhaj,
Liq. Cryst., 1995, 19, 537.
32 T. Mukai, M. Yoshio, T. Kato, M. Yoshizawaa and H. Ohn,
Chem. Commun., 2005, 1333.
48 A. Omenat, J. Barbera
´
,
J. L. Serrano, S. Houbrechts and
A. Persoons, Adv. Mater., 1999, 11, 1292.
49 C. Goltner, D. Pressner, K. Mullen and H. W. Spiess, Angew.
Chem., Int. Ed. Engl., 1993, 32, 1660.
¨
¨
33 H. Shimura, M. Yoshio, A. Hamasaki, T. Mukai, H. Ohno and
T. Kato, Adv. Mater., 2009, 21, 1591.
50 I. Miglioli, L. Muccioli, S. Orlandi, M. Ricci, R. Berardi and
C. Zannoni, Theor. Chem. Acc., 2007, 118, 203.
34 M. Yoshio, T. Ichikawa, H. Shimura, T. Kagata, A. Hamasaki,
T. Mukai, H. Ohno and T. Kato, Bull. Chem. Soc. Jpn., 2007, 80, 1836.
35 J. Motoyanagi, T. Fukushima and T. Aida, Chem. Commun., 2005,
101.
36 J. L. Anderson, R. Ding, A. Ellern and D. W. Armstrong, J. Am.
Chem. Soc., 2005, 127, 593.
51 D. J. Gardiner, C. J. Davenport, J. Newton and H. J. Coles,
J. Appl. Phys., 2006, 99, 113517.
52 D. J. Gardiner and H. J. Coles, J. Phys. D: Appl. Phys., 2007,
40, 977; H. J. Coles and M. N. Pivnenko, Nature, 2005, 436, 997.
53 S. M. Morris, M. J. Clarke, A. E. Blatch and H. J. Coles, Phys.
Rev. E, 2007, 75, 041701.
37 X. X. Han and D. W. Armstrong, Org. Lett., 2005, 7, 4205.
38 K. Huang, X. X. Han and X. T. Zhang, Anal. Bioanal. Chem.,
2007, 389, 2265.
39 J. L. Anderdson and D. W. Armstrong, Anal. Chem., 2005,
77, 6453.
40 S. Kumar and S. K. Varsheny, Org. Lett., 2002, 4, 157.
41 L. Zhang, H. Gopee, D. L. Hughes and A. N. Cammidge, Chem.
Commun., 2010, 46, 4255.
54 (a) X. Li, D. W. Bruce and J. M. Shreeve, J. Mater. Chem., 2009,
19, 8232; (b) J. E. Bara, E. S. Hatakeyama, B. R. Wiesenauer,
X. H. Zeng, R. D. Noble and D. L. Gin, Liq. Cryst., 2010,
37, 1587.
55 B. Santanu and D. Soma, J. Chem. Soc., Chem. Commun., 1995,
651.
56 S. De, V. K. Aswal, P. S. Goyal and S. Bhattacharya, J. Phys.
Chem., 1996, 100, 11664.
42 L. L. Li, P. Hu, B. Q. Wang, W. H. Yu, Y. Shimizu and
K. Q. Zhao, Liq. Cryst., 2010, 37, 499.
57 M. Dreja, S. Gramberg and B. Tieke, Chem. Commun., 1998, 1371.
58 S. Kumar and S. K. Gupta, Tetrahedron Lett., 2010, 51, 5459.
59 A. E. Bradley, C. Hardacre, J. D. Holbrey, S. Johnston, S. E. J.
McMath and M. Nieuwenhuyzen, Chem. Mater., 2002, 14, 629.
60 M. A. Alam, J. Motoyanagi, Y. Yamamoto, T. Fukushima,
J. Kim, K. Kato, M. Takata, A. Saeki, S. Seki, S. Tagawa and
T. Aida, J. Am. Chem. Soc., 2009, 131, 17722.
61 C. P. Fredlake, J. M. Crosthwaite, D. G. Hert, S. N. Aki and
J. F. Brennecke, J. Chem. Eng. Data, 2004, 49, 954.
62 J. Yang, J. F. Ma, Y. Y. Liu, S. L. Li and G. L. Zheng, Eur. J.
Inorg. Chem., 2005, 2174.
43 P. Herwig, C. W. Kayser, K. Mullen and H. W. Spiess, Adv. Mater.,
1996, 8, 510.
¨
44 C. T. Imrie and P. A. Henderson, Chem. Soc. Rev., 2007, 36, 2096.
45 T. Donaldson, H. Staesche, Z. B. Lu, P. A. Henderson,
M. F. Achard and C. T. Imrie, Liq. Cryst., 2010, 37, 1097.
46 N. Boden, R. J. Bushby, A. N. Cammidge, A. El-Mansoury,
P. S. Martin and Z. B. Lu, J. Mater. Chem., 1999, 9, 1391.
47 X. F. Kong, Z. Q. He, Y. N. Zhang, L. P. Mu, C. J. Liang,
B. Chen, X. P. Jing and A. N. Cammidge, Org. Lett., 2011, 13, 764.
c
2918 New J. Chem., 2011, 35, 2910–2918
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011